Question

I’m reading Eckel’s book, IO chapter, and there is the following code (p. 667).

public static void main(String[] args) throws IOException {
    try {
        DataInputStream in = new DataInputStream(new ByteArrayInputStream(BufferedInputFile.read("src/io/FormattedMemoryInput.java").getBytes()));
        while(true) {
            System.out.print((char)in.readByte()); // problem line
        }
    } catch (EOFException ex) {
        System.err.println("End of stream");
    }
}

This code works great, but if i change (char) in.readByte() to in.readChar() it prints me some asian symbols 灡捫慧攠楯㬊੩浰潲琠橡癡⹩漮⨻੩浰. Why is that and why it doesn’t print english ASCII symbols out?

Answer 1

Why is that and why it doesn’t print english ASCII symbols out?

From DataInput.readChar():

Reads two input bytes and returns a char value. Let a be the first byte read and b be the second byte. The value returned is:
(char)((a << 8) | (b & 0xff))

This method is suitable for reading bytes written by the writeChar method of interface DataOutput.

In other words, it’s treating your file as if it’s UTF-16-encoded – and it almost certainly isn’t.

When you want to read text data you should use a Reader subclass, e.g. InputStreamReader wrapped around FileInputStream, specifying the appropriate encoding for the input data.